Loading ...
Sorry, an error occurred while loading the content.

CMMi vs TMMi

Expand Messages
  • jyothirmayi mudumba
    Greetings,   Request the group to educate me on the below lines.   We are CMMi Level5 Organisation.And we are into TESTING services since 2 yrs and we had
    Message 1 of 11 , Nov 1, 2009
    • 0 Attachment

      Greetings,
       
      Request the group to educate me on the below lines.
       
      We are CMMi Level5 Organisation.And we are into TESTING services since 2 yrs and we had defined the testing processes inline with CMMi.
       
      In this senario going forward as we cld see a  good growth  in testing want to know is this the rite time and feasiable to have assesment in TMMi.
       
      OR
       
      By revisiting the and defining the process as per CMMi to the testing services inline with the bussiness growth is the good thought.
       
      Kidly


      Get your preferred Email name!
      Now you can @... and @....
    • EDWARD F WELLER III
      I guess the first question I would have is : Are you doing full life cycle testing services - that is, from the user/customer requirements, or a sub set of
      Message 2 of 11 , Nov 1, 2009
      • 0 Attachment
        I guess the first question I would have is : Are you doing full life cycle testing services - that is, from the user/customer requirements, or a sub set of this, where the other extreme is your customer provides the tests, you execute them and report results back.
         
        If the former, then you are doing test requirements analysis, test design, test verification (are the tests designed/implemented correctly) and then product verification. IN this case all the engineering Process Areas would apply to the work you do.
         
        If the later, then you are essentially a service organization and the SVC model may be a better fit.
         
        The TMMI model has only two accredited assessment/appraisal organizations (Google TMMI for information) and goes through "Level 3" so From my perspective is this a route that is accepted in the industry and will have a long term and wide influence in the industry?
         
        You also have to decide whether or not there is sufficient value add to your test processes if you are currently appraising your test projects as you mention - if they are CMMI L5, will TMMI L3 give you any advantage other than another marketing claim?
         
        Hope this helps
         
        Ed 
        ----- Original Message -----
        Sent: Sunday, November 01, 2009 4:16 AM
        Subject: [CMMi Process Improvement] CMMi vs TMMi

         


        Greetings,
         
        Request the group to educate me on the below lines.
         
        We are CMMi Level5 Organisation. And we are into TESTING services since 2 yrs and we had defined the testing processes inline with CMMi.
         
        In this senario going forward as we cld see a  good growth  in testing want to know is this the rite time and feasiable to have assesment in TMMi.
         
        OR
         
        By revisiting the and defining the process as per CMMi to the testing services inline with the bussiness growth is the good thought.
         
        Kidly


        Get your preferred Email name!
        Now you can @... and @rocketmail. com.

      • Buglione Luigi
        Hi Kidly, IMHO you could join the two models, reinforcing CMMI-DEV (that s what I call an horizontal model, following the whole lifecycle) with specific TMMi
        Message 3 of 11 , Nov 2, 2009
        • 0 Attachment
          Hi Kidly,
           
          IMHO you could join the two models, reinforcing CMMI-DEV (that's what I call an 'horizontal' model, following the whole lifecycle) with specific TMMi (that's what I call a 'vertical' model, specifying and making deeper analysis on a particular process or process group) PAs.
          You could take advantages also from a reduction of the number (and cost) for internal audits, since you'd enlarge the scope of your internal BPM (Business Process Model).
           
          Could it be feasible for your organization?
           
          Best regards
          Luigi
          ____________________________________________
          Luigi Buglione, Ph.D., IFPUG CSMS Gold Level
          SEMQ:     www.semq.eu
          ____________________________________________


          From: EDWARD F WELLER III [mailto:edwardfwelleriii@...]
          Sent: domenica 1 novembre 2009 16.40
          To: cmmi_process_improvement@yahoogroups.com
          Subject: Re: [CMMi Process Improvement] CMMi vs TMMi

           

          I guess the first question I would have is : Are you doing full life cycle testing services - that is, from the user/customer requirements, or a sub set of this, where the other extreme is your customer provides the tests, you execute them and report results back.
           
          If the former, then you are doing test requirements analysis, test design, test verification (are the tests designed/implemente d correctly) and then product verification. IN this case all the engineering Process Areas would apply to the work you do.
           
          If the later, then you are essentially a service organization and the SVC model may be a better fit.
           
          The TMMI model has only two accredited assessment/appraisa l organizations (Google TMMI for information) and goes through "Level 3" so From my perspective is this a route that is accepted in the industry and will have a long term and wide influence in the industry?
           
          You also have to decide whether or not there is sufficient value add to your test processes if you are currently appraising your test projects as you mention - if they are CMMI L5, will TMMI L3 give you any advantage other than another marketing claim?
           
          Hope this helps
           
          Ed 
          ----- Original Message -----
          Sent: Sunday, November 01, 2009 4:16 AM
          Subject: [CMMi Process Improvement] CMMi vs TMMi

           


          Greetings,
           
          Request the group to educate me on the below lines.
           
          We are CMMi Level5 Organisation. And we are into TESTING services since 2 yrs and we had defined the testing processes inline with CMMi.
           
          In this senario going forward as we cld see a  good growth  in testing want to know is this the rite time and feasiable to have assesment in TMMi.
           
          OR
           
          By revisiting the and defining the process as per CMMi to the testing services inline with the bussiness growth is the good thought.
           
          Kidly


          Get your preferred Email name!
          Now you can @... and @rocketmail. com.

        • Thomas George
          Hi, Given the Total Effort to develop a sftware product (including QA) and the Total Number of Filed Defects (Tickets) for that product, can we define
          Message 4 of 11 , Nov 2, 2009
          • 0 Attachment
            Hi,
             
            Given the Total Effort to develop a sftware product (including QA) and the Total Number of Filed Defects (Tickets) for that product, can we define a Quality measure to asses the Quality of that product?
             
            I hope my question is clear. Let me know, if you have any other thoughts on defining a Quality measure given these 2 data points.
             
            Thanks in advance!
             
            Regards, George.

          • Chris Simon
            Hello George,   My thoughts are that, with your first metric- effort to develop a product, you can measure efficiency or productivity and might not reflect
            Message 5 of 11 , Nov 3, 2009
            • 0 Attachment
              Hello George,
               
              My thoughts are that, with your first metric- effort to develop a product, you can measure efficiency or productivity and might not reflect directly on quality of product. It is very handy when it comes to planning 
              However, the second metric- # of defects (intentionally avoided "filed") can be a measure of quality. In the case of number of defects, one metric that comes to mind if FTNR (FTR can be used based on context). Defect density, defect deviations (variance factor), risk impact, RCA attribution etc are few derived metrics from # of defects which can have further analytical connotations. There are host of statistical analysis that can be done on the data around defects. Note, the rigor in analysis should make business sense primarily.
              Having said that, defect definition becomes critical and so does detection. At the end of the day, what matters is that the product (software in your case) should meet established requirements set by the end user, your business or customer in terms of functionality, product specifications & standards et al and at a higher level meeting the objectives/purpose/intent of the software
               
              Regards
              Chris

              --- On Tue, 3/11/09, Thomas George <thomas_george@...> wrote:

              From: Thomas George <thomas_george@...>
              Subject: [CMMi Process Improvement] Quality measure of a Software Product (Release)
              To: cmmi_process_improvement@yahoogroups.com
              Date: Tuesday, 3 November, 2009, 12:14 PM

               
              Hi,
               
              Given the Total Effort to develop a sftware product (including QA) and the Total Number of Filed Defects (Tickets) for that product, can we define a Quality measure to asses the Quality of that product?
               
              I hope my question is clear. Let me know, if you have any other thoughts on defining a Quality measure given these 2 data points.
               
              Thanks in advance!
               
              Regards, George.



              Add whatever you love to the Yahoo! India homepage. Try now!
            • Bruce R. Duncil
              Hello George, Obviously what you are already measuring doesn t meet your information needs and your objectives. First define your goals and objectives, what
              Message 6 of 11 , Nov 3, 2009
              • 0 Attachment
                Hello George,
                Obviously what you are already measuring doesn't meet your
                information needs and your objectives.

                First define your goals and objectives, what you are trying to
                achieve. Determine which questions those goals require that you
                answer. Those information needs must then drive what measures you
                need to collect and analyze. Measurement programs rarely if ever work
                in reverse.

                In this case, you want to understand and improve the quality of your
                product. Why? What is the impact of your current product quality?
                Who's affected? What do you need to know to define product quality?
                What is your total cost of poor quality? What measures are required
                to answer those questions? Do you have a breakdown of effort and cost
                for preventive and corrective quality measures you already take? How
                about the ones you need to take? Do you have a dollar figure for
                finding (or your "client" finding) and fixing your tickets? These are
                just some of the questions you should consider answering as part of
                defining a viable metrics program.

                Getting professional help with establishing your measurement program
                would save you a lot of wasted time and effort in helping your
                organization achieve its goals.

                Best Regards,
                Bruce
                www.alderonconsulting.com



                At 01:44 AM 11/3/2009, you wrote:
                >
                >
                >Hi,
                >
                >Given the Total Effort to develop a sftware product (including QA)
                >and the Total Number of Filed Defects (Tickets) for that product,
                >can we define a Quality measure to asses the Quality of that product?
                >
                >I hope my question is clear. Let me know, if you have any other
                >thoughts on defining a Quality measure given these 2 data points.
                >
                >Thanks in advance!
                >
                >Regards, George.
                >
                >
              • rob.leinen
                George, I don t believe so. I have seen some folks report defects per hour as their quality measure. And although there is a correlation between hours and
                Message 7 of 11 , Nov 3, 2009
                • 0 Attachment
                  George,

                  I don't believe so. I have seen some folks report defects per hour as their quality measure. And although there is a correlation between hours and defects, there is also a reverse correlation between defects and hours. And they are not the same. The problem is that defects increase the effort/hours on a project, however defect age ultimately determines the effort impact of each defect. Obviously, defects introduced in requirements but not identified until testing will increase the project effort much more than say a defect introduce in development and caught in testing. What this means for a quality measurement based on defects per hour is that the defect count could be on the rise, while the defects per hour on the decline (because older defects generally add significantly more effort to the project) giving the false impression that the organization is improving from a quality perspective.

                  The better measurement would be defects per element of size.

                  Cheers,
                  Rob L.


                  --- In cmmi_process_improvement@yahoogroups.com, Thomas George <thomas_george@...> wrote:
                  >
                  > Hi,
                  >
                  > Given the Total Effort to develop a sftware product (including QA) and the Total Number of Filed Defects (Tickets) for that product, can we define a Quality measure to asses the Quality of that product?
                  >
                  > I hope my question is clear. Let me know, if you have any other thoughts on defining a Quality measure given these 2 data points.
                  >
                  > Thanks in advance!
                  >
                  > Regards, George.
                  >
                • KUMAR RAHUL
                  Hi, Can we have the figure for the total effort required to solve the Total Number of Filed Defects (Tickets) ? If we can have it, then we will can have
                  Message 8 of 11 , Nov 4, 2009
                  • 0 Attachment
                    Hi,


                    Can we have the figure for the total effort required to solve the "Total Number of Filed Defects (Tickets)"? If we can have it, then we will can have (Total effort required to solve the ticket/Total effort required to develop the product) as a ratio.

                    If this tends to "1", the product quality can be termed as low.

                    This is just a thought!!!



                    On Tue, Nov 3, 2009 at 12:14 PM, Thomas George <thomas_george@...> wrote:
                     

                    Hi,
                     
                    Given the Total Effort to develop a sftware product (including QA) and the Total Number of Filed Defects (Tickets) for that product, can we define a Quality measure to asses the Quality of that product?
                     
                    I hope my question is clear. Let me know, if you have any other thoughts on defining a Quality measure given these 2 data points.
                     
                    Thanks in advance!
                     
                    Regards, George.




                    --
                    Regards,
                    Rahul
                    Sent from Karnataka, India
                  • efwelleraol
                    This indicator would assume effort for resolving all tickets is properly collected. One of the hardest tasks in getting a true cost of non-conformance is
                    Message 9 of 11 , Nov 4, 2009
                    • 0 Attachment
                      This indicator would assume effort for resolving all tickets is properly collected. One of the hardest tasks in getting a true cost of non-conformance is proper recording of effort spent by developers when debugging/fixing/retesting. If not recorded in the proper bucket, then this or any measure will be misleading (worthless)

                      All too often (if there is even a set of buckets for different tasks) developers record time as "coding"

                      However if you have this discipline in place, then you will get a ratio of development to rework cost. If you are limiting this to production tickets, then the ratio should be very low (assuming 99%+ defect removal effectiveness prior to ship)

                      --- In cmmi_process_improvement@yahoogroups.com, KUMAR RAHUL <rahul235@...> wrote:
                      >
                      > Hi,
                      >
                      >
                      > Can we have the figure for the total effort required to solve the "Total
                      > Number of Filed Defects (Tickets)"? If we can have it, then we will can have
                      > (Total effort required to solve the ticket/Total effort required to develop
                      > the product) as a ratio.
                      >
                      > If this tends to "1", the product quality can be termed as low.
                      >
                      > This is just a thought!!!
                      >
                      >
                      >
                      > On Tue, Nov 3, 2009 at 12:14 PM, Thomas George <thomas_george@...>wrote:
                      >
                      > >
                      > >
                      > > Hi,
                      > >
                      > > Given the *Total Effort *to develop a sftware product (including QA) and
                      > > the *Total Number of Filed Defects (Tickets) *for that product, can we
                      > > define a Quality measure to asses the Quality of that product?
                      > >
                      > > I hope my question is clear. Let me know, if you have any other thoughts on
                      > > defining a Quality measure given these 2 data points.
                      > >
                      > > Thanks in advance!
                      > >
                      > > Regards, George.
                      > >
                      > >
                      > >
                      >
                      >
                      >
                      > --
                      > Regards,
                      > Rahul
                      > Sent from Karnataka, India
                      >
                    Your message has been successfully submitted and would be delivered to recipients shortly.